PE firms can use data science to delve deeper into the data provided by specific targets and combine these insights with external data sources. With advances in technology and processes, these insights can now be gained at deal speed. This will help PE firms build a more comprehensive and accurate picture of their target’s growth opportunities and potential performance improvements.
By using data science, PE firms can uncover potential value levers related to identified deals that might not be obvious in order to make more competitive deal decisions.
Identifying growth opportunities pre deal through data science
Establishing a winning valuation requires a clear picture of the threats and opportunities of a target’s business model.
- What is the target’s position among its competitors in the market?
- Are the sales, marketing, and customer success functions operating efficiently and effectively?
- What products, value propositions, and brands are underleveraged in the market?
These questions and many others are key to understanding potential growth opportunities. While traditional due diligence analysis tries to answer these questions at a high level, applying data science can quickly identify a range of hidden market signals and more accurately predict how specific actions or events will alter growth forecasts. Uncovering these insights can provide a definitive advantage in competitive bid situations. Data science can be used to identify a wide range of potential growth opportunities.
Examples of growth levers in action
Market opportunities: Market opportunities can be identified and vetted on a multidimensional scale by combining data sets that capture broad-based views of a particular market. For example, internal data at a retail or hospitality company is useful for identifying the “what” and “where” of a customer purchase, but that data does not tell the whole story. Combining external market data, such as consumer sentiment, that explains “why” a customer made a particular purchase, along with geospatial data, to identify where a potential customer may be open to messaging and communication, helps build a precise picture of customer opportunity. Data science can make those connections— painting a more detailed and accurate picture of a target’s growth prospects within the broader dynamics of its market.
Enhanced customer retention: A broad, stable, and loyal customer base contributes to the attractiveness of a potential target. A primary goal for PE firms post acquisition is often to improve customer retention. Using data science during the pre deal phase, firms can determine the expected odds of successfully retaining a target’s customer base by assessing a range of indicators, such as cross-channel sales, repeat customers, customer service tickets, and social media sentiment. These insights can alter the valuation and bid price—and potentially allow the PE firm to win a competitive bid based on better customer retention models.
Sales pipeline management: With access to a target’s existing sales system, PE firms can use data science to assess the risks and opportunities in the current sales pipeline–and identify patterns of successful sales representative activity over time. When combining information from the sales system with external data sets such as geospatial proximity data, economic trends, and marketing activity, PE firms can identify potential unseen opportunities or threats that can be incorporated into their valuation. These insights are simply not available through traditional data analysis processes.
Example: Justifying a winning bid by identifying white space growth opportunities pre deal
Challenge–A PE firm was considering the acquisition of a national QSR chain in the U.S. The PE firm believed the target company had strong fundamentals (e.g., consumer sentiment, growing market share, and profitable growth), which could be used to expand its national footprint and existing footprint white space. This was key to its investment value creation thesis.
Actions–Deployed an advanced M&A analytics platform to ingest 500 GB of data to analyze and reconcile store-level P&Ls within 24 hours.
- Data scientists blended internal data with external alternative data and developed regression and predictive models to understand profitability of existing and planned locations based on customer demographics, population concentration, store hours, and other factors.
- A natural language processing application was used to gather 20 million social media mentions related to the target and its competitors in order to conduct in-depth customer sentiment analysis based on period and location.
Result–PE firm won a competitive auction with a bid supported by tangible and immediate growth opportunities identified during a rapid assessment conducted over a 72-hour period. The review of expansion activities highlighted potential locations with limited expected profitability, allowing management to redirect expansion to more profitable locations. A series of additional priority markets was also identified based on revenue potential and location characteristics (e.g., customer demographics and buying preferences). In the end, the analysis helped validate the potential white space growth prospects, which helped justify the valuation of the winning bid.
Projecting potential performance improvements
Uncovering opportunities to improve and sustain performance for the target can increase the valuation and lead to a winning bid.
- How do we align the operating model to better serve our business model?
- What processes could be adjusted to drive better efficiencies and reduce costs?
- How does the target’s technology infrastructure, operating model, and other factors compare to industry standards, and how can they be improved?
Data science is critical in helping PE firms identify opportunities to improve operational performance within a target company and the potential value associated with performance improvements across a broad variety of metrics. Analysis can be done quickly across a wide range of potential operational metrics.
Examples of operational levers in action
SKU rationalization: In the past, companies would often look at their “least profitable” or “negative profitable” SKUs as targets for elimination. However, data science offers a more refined way to develop rationalization models. Basic performance analysis can be augmented with halo and cannibalization effects, linear optimization, and forecasting to understand how products interact with each other and what the “real” effect on profitability is when adding to or removing from a portfolio. Sometimes customers who buy multiple SKUs with a company will stay with that company and buy a substitute SKU if one is removed. Other times, they will leave the company entirely. Understanding the difference is paramount, and data science can help.
Warehouse optimization: Another lever that companies can pull to reduce cost is through the consolidation and optimization of assets such as warehouses. Warehouse optimization allows for realignment of the number of warehouses with capacities while covering the demanding locations within a maximum given distance. Here, advanced analytics, in the form of footprint and sensitivity analytics, is used to identify which warehouses are operating sub optimally and need to be moved, possible places for relocation, and the net effect of that movement. This enables companies to reduce costs and, at the same time, improve the efficiency of their operations.
Example: Identifying performance improvement opportunities through data science to win a competitive bid
Challenge–A PE firm was seeking to acquire a restaurant franchisee with 1,100 units in the U.S. The valuation thesis required improving store performance through a more efficient delivery business. To support the valuation, the PE firm needed to diagnose the underlying drivers of a four wall revenue and cost—and identify potential improvement opportunities.
Actions–By combining internal data sets such as monthly P&L with external data sets including regional demographic, competitor benchmarks, geospatial data and more, machine learning models were developed to identify characteristics correlated with strong financial performance.
Using these insights, the PE firm was able to identify the optimal levers that would improve store performance.
Result–At deal speed, the firm was able to develop a plan to improve EBITDA by $30–$40 million per year through specific levers of performance improvement. The firm also found opportunities to reduce area manager count by one-third while keeping travel times consistent. The analysis validated the deal hypothesis and led to a successful bid in a competitive context.